Entropy最新文献

In this paper, we propose a method, namely Goalie, to defend against the correlated value and sign encoding attacks used to steal shared data from data trusts. Existing methods prevent these attacks by perturbing model parameters, gradients, or training data while significantly degrading model performance. To guarantee the performance of the benign models, Goalie detects the malicious models and stops their training. The key insight of detection is that encoding additional information in model parameters through regularization terms changes the parameter distributions. Our theoretical analysis suggests that the regularization terms lead to the differences in parameter distributions between benign and malicious models. According to the analysis, Goalie extracts features from the parameters in the early training epochs of the models and uses these features to detect malicious models. The experimental results show the high effectiveness and efficiency of Goalie. The accuracy of Goalie in detecting the models with one regularization term is more than 0.9, and Goalie has high performance in some extreme situations. Meanwhile, Goalie takes only 1.1 ms to detect a model using the features extracted from the first 30 training epochs.

In coal chemical industries, the optimal allocation of gas and steam is crucial for enhancing production efficiency and maximizing economic returns. This paper proposes an optimal scheduling method using operating zone models and entropy weights for an energy system in a gas-to-methanol process. The first step is to develop mechanistic models for the main facilities in methanol production, namely desulfurization, air separation, syngas compressors, and steam boilers. A genetic algorithm is employed to estimate the unknown parameters of the models. These models are grounded in physical mechanisms such as energy conservation, mass conservation, and thermodynamic laws. A multi-objective optimization problem is formulated, with the objectives of minimizing gas loss, steam loss, and operating costs. The required operating constraints include equipment capacities, energy balance, and energy coupling relationships. The entropy weights are then employed to convert this problem into a single-objective optimization problem. The second step is to solve the optimization problem based on an operating zone model, which describes a high-dimensional geometric space consisting of all steady-state data points that satisfy the operation constraints. By projecting the operating zone model on the decision variable plane, an optimal scheduling solution is obtained in a visual manner with contour lines and auxiliary lines. Case studies based on Aspen Hysys are used to support and validate the effectiveness of the proposed method.

This paper introduces DeeWaNA, an unsupervised network representation learning framework that unifies random walk strategies and neighborhood aggregation mechanisms to improve node classification performance. Unlike existing methods that treat these two paradigms separately, our approach integrates them into a cohesive model, addressing limitations in structural feature extraction and neighborhood relationship modeling. DeeWaNA first leverages DeepWalk to capture global structural information and then employs an attention-based weighting mechanism to refine neighborhood relationships through a novel distance metric. Finally, a weighted aggregation operator fuses these representations into a unified low-dimensional space. By bridging the gap between random-walk-based and neural-network-based techniques, our framework enhances representation quality and improves classification accuracy. Extensive evaluations on real-world networks demonstrate that DeeWaNA outperforms four widely used unsupervised network representation learning methods, underscoring its effectiveness and broader applicability.

The Dupire formula is a very useful tool for pricing financial derivatives. This paper is dedicated to deriving the aforementioned formula for the European call option in the space of distributions by applying a mathematically rigorous approach developed in our previous paper concerning the case of the Margrabe option. We assume that the underlying asset is described by the Merton jump-diffusion model. Using this stochastic process allows us to take into account jumps in the price of the considered asset. Moreover, we assume that the instantaneous interest rate follows the Merton model (1973). Therefore, in contrast to the models combining a constant interest rate and a continuous underlying asset price process, frequently observed in the literature, applying both stochastic processes could accurately reflect financial market behaviour. Moreover, we illustrate the possibility of using the minimal entropy martingale measure as the risk-neutral measure in our approach.

The rise in generative large models has gradually influenced traditional product design processes, with AI-generated content (AIGC) playing an increasingly significant role. Globally, tourism IP cultural products are crucial for promoting sustainable tourism development. However, there is a lack of practical design methodologies incorporating generative large models for tourism IP cultural products. Therefore, this study proposes a methodology for the color matching and shape design of tourism IP cultural products using multimodal generative large models. The process includes four phases, as follows: (1) GPT-4o is used to explore visitors' emotional needs and identify target imagery; (2) Midjourney generates shape options that align with the target imagery, and the optimal shape is selected through quadratic curvature entropy method based on shape curves; (3) Midjourney generates colored images reflecting the target imagery, and representative colors are selected using AHP and OpenCV; and (4) color harmony calculations are used to identify the best color combination. These alternatives are evaluated quantitatively and qualitatively using a color-matching aesthetic measurement formula and a sensibility questionnaire. The effectiveness of the methodology is demonstrated through a case study on the harbor seal, showing a strong correlation between quantitative and qualitative evaluations, confirming its effectiveness in tourism IP product design.

The goal of this research is to introduce and thoroughly investigate a new methodology for the assessment of sequential regularity in volatility time series. Three volatility estimators based on daily range data are analyzed: (1) the Parkinson estimator, (2) the Garman-Klass estimator, and (3) the Rogers-Satchell estimator. To measure the level of complexity of time series, the modified Shannon entropy based on symbol-sequence histograms is utilized. Discretization of the time series of volatility changes into a sequence of symbols is performed using a novel encoding procedure with two thresholds. Five main stock market indexes are analyzed. The whole sample covers the period from January 2017 to December 2023 (seven years). To check the robustness of our empirical findings, two sub-samples of equal length are investigated: (1) the pre-COVID-19 period from January 2017 to February 2020 and (2) the COVID-19 pandemic period from March 2020 to April 2023. An additional formal statistical analysis of the symbol-sequence histograms is conducted. The empirical results for all volatility estimators and stock market indexes are homogeneous and confirm that the level of regularity (in terms of sequential patterns) in the time series of daily volatility changes is high, independently of the choice of sample period. These results are important for academics and practitioners since the existence of regularity in the time series of volatility changes implies the possibility of volatility prediction.

Accurately determining the exposure time to an infectious pathogen, together with the corresponding incubation period, is vital for identifying infection sources and implementing targeted public health interventions. However, real-world outbreak data often include outliers-namely, tertiary or subsequent infection cases not directly linked to the initial source-that complicate the estimation of exposure time. To address this challenge, we introduce a robust estimation framework based on a three-parameter Weibull distribution in which the location parameter naturally corresponds to the unknown exposure time. Our method employs a γ-divergence criterion-a robust generalization of the standard cross-entropy criterion-optimized via a tailored majorization-minimization (MM) algorithm designed to guarantee a monotonic decrease in the objective function despite the non-convexity typically present in robust formulations. Extensive Monte Carlo simulations demonstrate that our approach outperforms conventional estimation methods in terms of bias and mean squared error as well as in estimating the incubation period. Moreover, applications to real-world surveillance data on COVID-19 illustrate the practical advantages of the proposed method. These findings highlight the method's robustness and efficiency in scenarios where data contamination from secondary or tertiary infections is common, showing its potential value for early outbreak detection and rapid epidemiological response.

This study examines the thermal conduction resistance in anisotropic bodies using linear extended irreversible thermodynamics. The fulfilment of the Onsager Reciprocal Relations in anisotropic bodies, such as crystals, has been demonstrated. This fulfilment is achieved by incorporating Newton's heat transfer coefficients into the calculation of the entropy production rate. Furthermore, a basic principle for the transport of heat, similar to the Onsager-Fuoss formalism for the multicomponent diffusion at a constant temperature, was established. This work has the potential to be applied not just in the field of material science, but also to enhance our understanding of heat conduction in crystals. A novel formalism for heat transfer analogous to Onsager-Fuoss model for multicomponent diffusion was developed. It is believed that this work could be applied for educational purposes.

Entropic Dynamics (ED) provides a framework that allows the reconstruction of the formalism of quantum mechanics by insisting on ontological and epistemic clarity and adopting entropic methods and information geometry. Our present goal is to extend the ED framework to account for spin. The result is a realist ψ-epistemic model in which the ontology consists of a particle described by a definite position plus a discrete variable that describes Pauli's peculiar two-valuedness. The resulting dynamics of probabilities is, as might be expected, described by the Pauli equation. What may be unexpected is that the generators of transformations-Hamiltonians and angular momenta, including spin, are all granted clear epistemic status. To the old question, 'what is spinning?' ED provides a crisp answer: nothing is spinning.

Groove, a complex aspect of music perception, plays a crucial role in eliciting emotional and physical responses from listeners. However, accurately quantifying and predicting groove remains challenging due to its intricate acoustic features. To address this, we propose a novel framework for groove rating that integrates Convolutional Neural Networks (CNNs) with Temporal Convolutional Networks (TCNs), enhanced by entropy regularization and entropy-pooling techniques. Our approach processes audio files into Mel-spectrograms, which are analyzed by a CNN for feature extraction and by a TCN to capture long-range temporal dependencies, enabling precise groove-level prediction. Experimental results show that our CNN-TCN framework significantly outperforms benchmark methods in predictive accuracy. The integration of entropy pooling and regularization is critical, with their omission leading to notable reductions in R² values. Our method also surpasses the performance of CNN and other machine-learning models, including long short-term memory (LSTM) networks and support vector machine (SVM) variants. This study establishes a strong foundation for the automated assessment of musical groove, with potential applications in music education, therapy, and composition. Future research will focus on expanding the dataset, enhancing model generalization, and exploring additional machine-learning techniques to further elucidate the factors influencing groove perception.